Few topics ignite more impassioned debate in the theatre tearoom than that of specialist obstetricians providing 24-hour labour ward cover. International practice varies; earlier this year, the American College of Obstetricians and Gynecologists endorsed the Obstetric Care Consensus, recommending on-site obstetrician cover for all Level III services, i.e. those providing care to women with complex maternal medical conditions, obstetric complications and fetal conditions. For many years, 24-hour specialist presence has also been mandatory for all North American centres offering residency programs. The Royal College of Obstetricians and Gynaecologists (RCOG) recommends a consultant-delivered service, particularly in units with high birth numbers (>5000/year), and the Royal Australian and New Zealand College of Obstetricians and Gynaecologists (RANZCOG) recommend that all training institutions must, as a minimum, have an on-call obstetric specialist who can attend within 30 minutes. A robust debate at the 2015 RCOG/RANZCOG annual scientific meeting revealed entrenched opinions regarding the potential adverse impact of 24-hour obstetric cover on job satisfaction, work–life balance and the ability to attract, and retain, new specialists. Yet voting for the status quo enshrines the myth that specialists being 30 minutes away from tertiary obstetric centres—which commonly face the combined challenges of high birth numbers, high case complexity and heavy training responsibilities—is always sufficient. Central to this debate is whether obstetric outcomes and training opportunities are improved by on-site specialists. Although these questions have not been addressed by controlled studies, other data provide circumstantial evidence that such benefits likely exist. A study of nearly 450 000 births reported an increased risk of intrapartum/early neonatal death at night than in daytime, both in women who required induction or augmentation of labour (OR 1.78; 95% CI 1.35–2.40), and in those requiring emergency caesarean section (OR 1.84; 95% CI 1.14–2.96), despite careful adjustment for likely confounding variables (Gijsen et al. BMC Pregnancy Childbirth 2012;12:92). Similar findings previously led to speculation that differences were attributable to less experienced caregivers at night, with most specialists ‘on-call from home…less likely to be involved in the judgement of (pre)critical conditions and the initial management of high-risk situations’ (deGraaf et al. BJOG 2010;117:1098–107). These findings contrast starkly with those from centres with on-site specialists. No association was observed between adverse perinatal outcome and time of delivery in a large Californian cohort (Caughey et al. Am J Obstet Gynecol 2008;199:496.e1–5), and another study of nearly 19 000 women requiring unscheduled caesarean delivery detected no increase in maternal morbidity or neonatal complications at night where 24-hour specialist presence was routine (Bailit et al. Am J Obstet Gynecol 2006;195:1132–7). These findings are unsurprising given that time-critical conditions tend to cluster, and synchronise, around tertiary centres: placental abruption, uterine rupture and other precipitants of acute maternal or fetal compromise where a 30-minute response time may prove inadequate. But surely this is how we all learned? Are historical models of training through trial and (occasional) error preferable, or is the training experience optimised in the North American model of consultant-led supervision? Certainly some time-critical events are appropriately managed by ‘solo’ trainees. Yet severe fetal compromise managed with an outlet ventouse is entirely different to that requiring caesarean where the patient's BMI is 50. This intersection between ‘time-critical’ and ‘experience-critical’ events is not infrequent, and is even more challenging given the diminished experience resulting from reduced training hours. Trials of forceps are now commonplace, yet successful vaginal birth is more likely when a specialist is present (Lewis et al. J Obstet Gynaecol 2011;31:229–31). This enforces the need for every training hour to count. Specialist presence can be expected to optimise the learning curve for achieving procedural competence, and improve assessment of knowledge, skills and attributes across the breadth of training domains. Twenty-four-hour cover is not feasible in all settings. Optimising remuneration and conditions is essential to ensure job satisfaction. Nevertheless, for those who birth, those who are born, and those who train in tertiary obstetric centres, 24-hour specialist presence should be considered an important step forward. None declared. Completed disclosure of interests form available to view online as supporting information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.